NFL Win Total Workbook: August 17th
The value of takeaways, and my takeaways about those takeaways.
This is the continuation of a summer-long series about NFL and CFB win total concepts. Part 1, with a simple NFL exercise to work on, can be found here. Parts 2 and 3, with an exercise and my thoughts on CFB win totals, can be found here and here. Part 4, about CFB schedule flexibility as a handicapping tool, can be found here.
I’m gonna be honest…there are some football analytics I don’t really have a great grasp of. How does Pro Football Focus come up with their grades for players? They never seem quite right. What is DVOA exactly? I kinda know, but I also kinda don’t. Expected-points-added (EPA) seems intuitive, but it’s still a little dense for a lot of people. A lot of people is me. I wrote that to make it seem like it wasn’t also me. It me.
There are things I think we all kinda get, though. And one of them is turnover margin.
Turnovers are bad when you are committing them. They are good when you are forcing them. When you force turnovers, you stop an opponent’s possession and you get a possession, usually in an advantageous part of the field vs. where you would normally gain possession. You score more points because you get more possessions. You win more. This is like caveman football. Take ball, score ball.
Here’s a chart to illustrate the value of turnover margin in a very basic way. This is every team that had a positive turnover margin last year, and the teams highlighted green made the playoffs.
Of the four teams listed who did not make the playoffs, Detroit and Green Bay both barely didn’t make it in the NFC (they famously played in the final game of the regular season with a playoff spot at stake for the Packers), and the Steelers and Patriots were the best two teams in the AFC record-wise that did not make the playoffs. You get the idea. Every one of these teams was “good” last year.
This doesn’t mean you HAVE to have a good turnover margin to win. Kansas City (-3) isn’t on this list. They were really good. Same with Buffalo (0, so I guess they were close). Working the stat the other way, Houston was only -1 last year in turnover margin, and they were basically the worst team in the league. But let’s just say if you can work the math to get this stat on your side, it’s a really, really good thing, that significantly increases your chances of winning.
There are obviously two parts to turnover margin: takeaways, and giveaways. Takeaways are how many times your defense can force turnovers (or how many times the offense you’re playing turns it over, if you think it’s more about the offense), and giveaways is how many times your team gives it to the other team, and that’s usually driven by your quarterback’s play. There’s obviously a luck component to this stat, and a skill component. Getting pressure off the edge, forcing a QB to fumble: skill! Recovering that fumble in a pile of 10 large humans: luck! A corner jumping a route and taking an interception the other way: skill! The ball hitting a ref and landing in a linebacker’s hands: luck!
Let’s get a little more into the weeds on turnover margin’s underlying aspects, how they can help us make predictions, and then what you can try to do to solve the puzzle of the 2023 NFL season.